tmux-agents
Skill1 starsChanged 3 months agoMITView on GitHub
Use when launching a multi-agent terminal workspace, pairing Claude Code with Google Antigravity (agy), managing Boss-Worker matrices, orchestrating Git worktrees, or capturing agent telemetry in tmux
In this repository65 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- async-python-patternsSkill
- bash-defensive-patternsSkill
- brainstormingSkill
- clean-systemSkill
- debian-linux-triageSkill
- dispatching-parallel-agentsSkill
- dotfilesSkill
- executing-plansSkill
- finishing-a-development-branchSkill
- handoffSkill
- harness-checkSkill
- mcp-builderSkill
- perf-tuneSkill
- prompt-architectSkill
- python-mcp-server-generatorSkill
- python-performance-optimizationSkill
- python-testing-patternsSkill
- receiving-code-reviewSkill
- requesting-code-reviewSkill
- security-and-hardeningSkill
- subagent-driven-developmentSkill
- sys-diagSkill
- systematic-debuggingSkill
- test-driven-developmentSkill
- tmux-agentsSkill
- update-runtimesSkill
- using-git-worktreesSkill
- using-superpowersSkill
- verification-before-completionSkill
- writing-plansSkill
- writing-skillsSkill
- wsl-snapshotSkill
- async-python-patternsSkill
- bash-defensive-patternsSkill
- brainstormingSkill
- clean-systemSkill
- debian-linux-triageSkill
- dispatching-parallel-agentsSkill
- dotfilesSkill
- executing-plansSkill
- finishing-a-development-branchSkill
- handoffSkill
- harness-checkSkill
- mcp-builderSkill
- perf-tuneSkill
- prompt-architectSkill
- python-mcp-server-generatorSkill
- python-performance-optimizationSkill
- python-testing-patternsSkill
- receiving-code-reviewSkill
- requesting-code-reviewSkill
- security-and-hardeningSkill
- subagent-driven-developmentSkill
- sys-diagSkill
- systematic-debuggingSkill
- test-driven-developmentSkill
- update-runtimesSkill
- using-git-worktreesSkill
- The first 60 shown.
The file
Files
Multi-Agent Tmux Session & Orchestration Skill
Orchestrates multi-pane terminal workflows pairing Claude Code with Google Antigravity (agy), Boss-Worker hierarchies, isolated Git Worktrees, and real-time telemetry in tmux 3.5a.
Trigger Scenarios
- Initializing collaborative dual-agent pairing (
agy+claude+ telemetry) - Launching scalable Boss-Worker agent companies (
companymode) - Creating isolated Git Worktrees for concurrent agent coding without file collision
- Capturing background buffer logs (
capture) for silent audits - Resetting context windows across all worker panes simultaneously (
clear-all) - Interactive agent workspace dashboard via floating popup (
prefix + A)
Invocation
${CLAUDE_PROJECT_DIR}/scripts/tmux_agents.sh [subcommand] [args]Subcommands & Options
| Subcommand | Arguments | Description |
start / pair | [session_name] | Initializes 3-pane paired agent session (agy, claude, btop/htop) or reattaches |
company | [n_workers] | Launches 1 Boss + N Worker panes in balanced tiled layout |
worktree add | <branch_name> | Spawns isolated Git worktree in .worktrees/<branch> with dedicated agent window |
worktree list | (none) | Lists all active agent worktrees |
worktree clean | <branch_name> | Safely removes worktree directory and branch |
capture | <target> [lines] | Silently grabs output buffer from target pane without terminal interruption |
clear-all | [session_name] | Sends /clear to all panes to reset token context |
status | (none) | Displays list of active agent sessions and panes |
kill | [session_name] | Gracefully terminates target agent session |
menu | (none) | Opens interactive TUI dashboard (bound to prefix + A) |
Safety Classification
- Tier 2 (Controlled System Operation): Authorized terminal workspace orchestrator managing user tmux sessions and sandboxed worktrees.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

